How I got my Microsoft certifications
Exams: AI-900, DP-900, PL-900, PL-300, DP-500
It has been 9 months now since I was hired by NTT Data as a Data Analyst, and in the last months, I spent a lot of time studying and preparing myself for Microsoft Certification exams.
As a Data Analyst, I started working with specific tools for data analysis, exploration and querying, data manipulation, and data visualization. However, I also wanted to put (and explore) the theoretical approach alongside the practical one, and this latter helped me pass the exams with a relatively small effort.
In this very young career of mine as a Data Analyst, I chose the following Microsoft certifications because they seemed to represent the most logical path to becoming a professional Data analyst:
- AI-900, Azure AI fundamentals: this certification is centered around the main and basic concepts related to AI and Machine Learning, along with Azure services and tools to put them into practice (or at least, to start using them).
- DP-900, Azure Data fundamentals: the basic concepts about databases and Azure services to work and manipulate data are the main arguments of this certification exam.
- PL-900, Power Platform fundamentals: along with the most known Power BI service, this certification concerns the basic concepts behind the Microsoft Power Platform, and how to use the main provided services to build apps, reports, flows, and bots.
- PL-300, Power BI Data Analyst Associate: the certification is about how to import, explore, manipulate, and visualize data using the main tools offered by Power BI. Moreover, it is also centered around the main functions of the DAX language to create ad-hoc measures.
- DP-500, Azure Enterprise Data Analyst Associate: the certification can be considered as an advanced PL-300 certification for what concerns some specific topics. Moreover, it explores how to design and implement enterprise-scale analytics solutions using Microsoft Azure (in particular, Microsoft Purview and Azure Synapse Analytics) and Power BI (Desktop and Service).
The certifications above are all related to a broad range of Data Analysis aspects that I encounter every day, which pushed me to take those exams.
Learning path & materials
In the first months in NTT Data, my knowledge of concepts and tools related to Data Analysis was very limited, even though I had all the necessary practical and mathematical bases to understand them. What I lack was the set of tools and concepts related to the Microsoft and the Azure world along with some notions that I have never encountered before.
In other words, in the first months as a Data Analysist, I was partially prepared only for the AI-900 exam (given my primary interest in Machine Learning) but I was completely unprepared for the others.
The reason was that I was not entirely confident about the other services (like Power BI and the other Power Platform services) and my knowledge was limited to what I have learned and explored in the first months.
The strategy I adopted was very simple: put theory and practice alongside to each other. Every time I found something new or different with respect to what I knew, I quickly tested it myself to learn by example. And this strategy was absolutely successful for all the exams I prepared.
I studied the theory using all the possible sources from the web, starting from the learning paths and virtual training days offered by Microsoft itself. All the notes I have taken in my notebooks were integrated and enriched with other information found on the Internet or by simply testing the topics directly on the main tools I was using (i.e., Power BI and Power Platform for reports, apps and flows, Virtual Studio Code for Machine Learning, Data analysis and Python). I also used Youtube where some people have uploaded several videos with useful overviews about the exams I wanted to take.
Questions & Tests
I must admit: at the beginning, I was completely lost. When I used the available free tests offered by Microsoft, I was also surprised by the fact that those questions were not real examples of what I would have encountered during the exam: later, I understood that those tests were useful to find possible weaknesses or simply to test the basic concepts explained in the online Microsoft learning materials.
Then I looked to the web and found a lot more (even people surprised as me for those very “easy” questions :D). I found that the real exam questions cover a broader area than the one contained in the Microsoft learning materials but all I needed was there, in the Microsoft documentation. Moreover, I found very useful websites with free-to-consult questions, real questions that happened to occur in the exams. The websites are the following:
- examtopics: you need to create an account and pay to explore all the questions in the database, but it is possible to study from the available questions and discussions.
- itexams: like the previous website, you need to pay to explore all the available questions.
- passnexam: very useful with a lot of available questions (not always “real” exam questions) but completely free to consult.
Also on Youtube, I found several videos (that I watched a few times) and people showing and explaining exam questions.
So, day by day, I started a series of steps that eventually became a true routine. Along with my job activities (occupying most of the day), I found a little bit of time to pursue my certification goals, and a few days ago a obtained the last certification (DP-500).
So, with theory, documentation, and questions, I had almost everything I needed to pass the exams. The last (but not least) important element was consistency. Every day, after work, I spent at least half an hour studying, trying tests, and exploring the Microsoft documentation when I found unclean questions. The routine I started was very simple but very powerful to pass all the exams (and possibly, the next ones :D). I can summarize it as the following:
- Go through the Microsoft materials (i.e., learning paths, virtual training days) and take notes about everything, even if some concepts are known.
- Review notes and enrich them with information coming from different sources (i.e., Internet, books) or even from personal experimentations with official services and tools.
- Try tests and study the questions not correctly answered to understand weaknesses.
- Read and participate in forums and discussions for more information. Very often, some people may provide you with useful documentation or comments that can help to understand a concept better.
- Always practice and explore topics “on the field”. Do not stop (never!) at the simple question-answer: always go beyond it to get all possible information. In my opinion, this was the key ingredient to passing all the exams: putting those exam topics into practice helped me understand unknown concepts and reinforce those I already knew.
I applied this routine almost every day (and keep on applying it also for other tasks) because it is the strongest way to reinforce your knowledge (see also my previous post How do I learn?)
For what concerns the exams themselves, I have the following comments:
- AI-900: most of the questions are theoretical with a few concerning practical applications of Machine Learning concepts in the Microsoft Machine Learning Studio. Theory may be enough to pass the exam, but if you want to continue on the Machine Learning certification path, practice is essential.
- DP-900: like the previous exam, most of the questions are theoretical and regard the large number of services and tools offered by Microsoft through the Azure cloud platform to work with data (e.g., Azure databases, Synapse Analytics, Data Factory, and many others). A deep dive into those services is crucial to understand them better (for this, I suggest this overview video on Youtube).
- PL-900: many questions are related to practical applications in the Power Platform services (mostly, PowerApps and Power Automate). For what concerns this exam, I attended a Microsoft virtual training day, useful to strengthen basic concepts, but not completely related to the real exam questions.
- PL-300: for me, this was a clear example of what the theory-practice approach can do. Studying and using Power BI was truly crucial to pass the exam. Questions are mainly related to data modeling, visualization, and transformation, along with questions concerning DAX syntax and measures.
- DP-500: if you pass the PL-300 exam, you have a good chance to pass this exam as well, given that many concepts are already covered in the previous exam course. Here a good knowledge of Python may be very useful. Moreover, many questions are related to Synapse Analytics and how to manage an enterprise-scale analytics solution.
So, my final words are these: first of all, mind that a certification does not necessarily coincide with complete expertise: this latter comes with application and consistency. Therefore, do not panic or feel frustrated when the learning path seems too harsh and difficult. Keep on going and be consistent with your routine, follow eagerly the theory-practice approach, and results will come all along! :D
